python读取mat数据集

以http://ufldl.stanford.edu/housenumbers/上的mat数据集为例
需要注意以下几点
从mat提取出来的数据以字典的形式保存，所以需要提取字典的key和value

import numpy as np
import os
from PIL import Image
import matplotlib.pyplot as plt
import scipy.io as scio

data_train=scio.loadmat(r'data/train_32x32.mat')
#print(data_train)
#取出需要的数据矩阵
data_train_data=data_train.get('X')
#print(data_train_data)

#输出字典里的X
#data_train_label=data_train.get('y')#提取标签
#print(len(data_train_label))
for i in range(10):
    plt.subplot(2,5,i+1)
    #plt.title(data_train_label[i][0])
    #print("标签"+str(data_train_label[i][0]))
    plt.imshow(data_train_data[...,i])
    #print("数据"+str(data_train_data[...,i])+"------")
    plt.axis('off')

plt.savefig('picture/1.png')
plt.show()

输出结果

批量处理
将mat文件里的所有图片进行输出并保存在不同的画布中

import numpy as np
import os
from PIL import Image
import matplotlib.pyplot as plt
import scipy.io as scio

data_train=scio.loadmat(r'data/train_32x32.mat')
print(data_train)
#取出需要的数据矩阵
data_train_data=data_train.get('X')
print(data_train_data)

#输出字典里的X
data_train_label=data_train.get('y')#提取标签
print(len(data_train_label))
for i in range(10):#如果要全部输出，这里应该改成73256
    #plt.subplot(1,1,i+1)
    #plt.figure()
    plt.title(data_train_label[i][0])
    print("标签"+str(data_train_label[i][0]))
    plt.imshow(data_train_data[...,i])
    print("数据"+str(data_train_data[...,i])+"------")
    plt.axis('off')
    plt.savefig('picture/' +str(i)+'.png')
plt.show()

#此函数读取特定文件夹下的mat格式图像，返回图片所在路径的列表
def get_imlist(path):
    for f in os.listdir(path):
        if f.endswith('.mat'):
            return os.path.join(path,f)#路径加起来
            #print(os.path.join(path,f))

c=get_imlist(path)
print(c)
d=len(path)
print(d)

输出结果